Near-Minimax-Optimal Risk-Sensitive Reinforcement Learning with CVaR.
Kaiwen WangNathan KallusWen SunPublished in: CoRR (2023)
Keyphrases
- risk sensitive
- optimal control
- risk neutral
- reinforcement learning
- model free
- markov decision processes
- dynamic programming
- risk averse
- optimality criterion
- control policies
- worst case
- utility function
- function approximation
- average cost
- reinforcement learning algorithms
- control strategy
- machine learning
- markov decision problems
- control policy
- pareto optimal
- optimal solution
- evaluation function
- game theory
- optimal policy
- multi agent
- decision makers